CompTIA Cloud+ CV0-003 Exam: Disaster Recovery Tasks

Objective: Given a scenario, perform disaster recovery tasks.

You will learn about failovers, failbacks, replication, and standby sites. You will also learn about the documentation involved in disaster recovery and the geographical requirements you should consider when choosing a data center.

Topics:
- Failovers
- Failback
- Restore Backups
- Replication
- On-Premises and Cloud Sites
- Requirements
- Documentation
- Geographical Data Center Requirements

1. When a(n) _____ system is available and an organization switches over to this system, this is called a failover.
2. In a(n) _____ standby site, the site is always available for immediate failover.
3. A(n) _____ includes all of the resources and tools needed to perform all of the operations of your disaster recovery plan.
4. True or false: The political climate for an area should not be a consideration when choosing the geographic location of a data center.

Answers:

1. Redundant
2. Hot
3. Disaster recovery (DR) kit
4. False

Failovers
Many organizations have redundant systems if one system fails. Some examples include
- Hardware redundancy in the event that a hardware device fails
- Application redundancy if an application crashes or is no longer available via the network

When a redundant system is available and an organization switches over to the redundant system, this is called a failover. In some cases, this failover may happen automatically. For example, a monitoring process may recognize when the original system is not responding and then invoke the failover process.
In most cases, once a failover occurs, the reason for the failure should be explored; this process is also known as a root cause analysis. Ideally, when the problem is discovered and fixed, the system should revert back to using the original.

Failback
After a failover occurs, it is important to restore the original system and revert the service back to the original system. This process is called failback.
It is important to implement a failback operation in a timely manner because normally the redundant system isn’t designed for long-term use. For example, it may have fewer hardware resources available. Think of the redundant system like a spare tire in a car; while the spare will work in a pinch, it isn’t normally designed to completely replace a regular tire. This isn’t always the case, but consider that redundant systems can double the costs for the resource if they are as fully functional as the original.

Restore Backups
For more information on restoring backups, see the “Restoration Methods” section in this guide.

Replication
We already discussed recovery point objectives (RPOs). This discussion was related to how important it was to have timely backups to ensure that an organization is able to meet its RPO objectives. That discussion covered only backing up data in a timely manner, which is important in situations when data is corrupted or lost.
But what if an entire instance or resource is lost? For example, what if a virtual machine that hosts a critical web server crashes and can’t be recovered? In these situations, you should have one or more redundant instances in place so that you can use a redundant instance as a fallback (see the earlier “Failback” section).
To have a valid redundant resource, you need to replicate the active resource often enough to support your RPO. For example, if that critical web server has an RPO of 6–12 hours, you must ensure you use a replication method that updates the redundant systems at least every 12 hours.
Note that some cloud vendors offer replication for some products automatically. Databases, for example, are often considered to be mission-critical resources. As a result, cloud vendors often provide an automatic replication option for database resources.

On-Premises and Cloud Sites
When dealing with mission-critical cloud-based resources, you may consider having a standby site available to quickly resume operations. This standby site can be defined as either hot, warm, or cold. In this section you’ll learn the difference between the three types.

Hot
In a hot standby site, the site is always available for immediate failover. This type of standby site provides the quickest availability and is used in situations in which downtime must be avoided at all costs. There are some disadvantages of a hot site:
- To replicate the active site as accurately as possible, the replication process may be complex and time-consuming to manage.
- The hot site itself must be managed, typically by a dedicated support team.
- The cost of maintaining a hot site is higher than maintaining warm or cold sites.

If the hot site is on-premises, the organization needs to provide all of the hardware components and the dedicated team to manage the site. Given that a standby site should be in a different geographic location than the active site to allow for recovery from natural disasters, an on-premises hot site can be very costly for an organization.
A cloud-based hot site is typically managed by a cloud vendor or partner. This type of hot site may result in lower costs if a facility that is shared by other clients is used. However, most organizations prefer a hot site to be physically separated from other clients, so the cost of a cloud hot site can also be high.

Warm
While a warm site isn’t immediately available for failover, it should be available fairly quickly. Given the complexity of the active site, this may be anywhere from hours to up to a day. Warm sites are more cost effective than hot sites and, for many situations, have a reasonable delay before the site is available. Because a warm site is not always active, the replication process may result in some lost data or application configuration. Also, a warm site isn’t normally available for testing, so it may not be as reliable as a failover as a hot site is.
A warm on-premises site may be hosted in the data center of a branch office for an organization. A branch office can be ideal because it should have IT staff on hand to handle the rest of that data center’s needs. Warm cloud standby sites are typically shared with other customers to reduce costs.

Cold
A cold standby site is one that needs to be provisioned before it is available. The hardware should be available, but backups must be used to restore services. This type of standby site has not only the lowest cost but also the highest downtimes.
Cold on-premises standby sites are rare because they require dedicated hardware to be always available. This hardware would just be sitting there, waiting for a disaster to occur. More likely, a cloud vendor would provide a disaster recovery team to aid an organization in recovering its site.

Requirements
Several requirements topics were covered in this guide. The information for the pertinent sections is provided in the following sections in that guide:
- RPO:See “Recovery Point Objective (RPO)”
- RTO:See “Recovery Time Objective (RTO)”
- SLA:See “Service-Level Agreements (SLAs)”

Corporate Guidelines
Most large organizations (and many smaller ones) have well-established corporate guidelines related to IT operations. This includes recovery rules that must be followed when establishing disaster recovery tasks.
Before developing any recovery policies, make sure you are aware of your organization’s corporate guidelines. This includes not only the procedures and rules you must follow when performing recovery operations but also what reports need to be created after the recovery process is complete.
Corporate guidelines will often also mandate how the recovery process is documented. See the following section for more details.

Documentation
In the middle of a disaster recovery is not the time to try to figure out what your organization needs to do to recover. Well-thought-out plans should be put in place, and every step of the recovery should be well documented. In this section you’ll explore some key components to disaster recovery documentation.
Note that this documentation is governed by the organization’s business continuity plan (BCP). The BCP is a document that describes how the organization will be able to function during a disaster or an emergency.

DR Kit
A disaster recovery (DR) kit includes all of the resources and tools needed to perform all of the operations of your disaster recovery plan. Exactly what is in the kit depends on your infrastructure, but it could include essential user names and passwords, contract information for key individuals, and software licenses.
In disaster recovery a playbook provides the step-by-step tasks that need to be performed to conduct the disaster recovery process. The playbook should be very specific and should make clear who is responsible for which tasks. The playbook should include many of the components discussed in this guide and the topics from here. This includes defining the RTOs, methods to monitor the recovery process, and how to utilize the DR kit.

Network Diagram
When you’re dealing with complex network structures, especially in the cloud, having a network diagram can be critical when recovering all or parts of a network structure. A network diagram illustrates where all of the different components of the network reside, including network configuration settings (IP address, subnet, and so on).
Many tools are available for creating network diagrams. For example, the below figure displays the Lucidchart online tool, which has a free mode available in the event that you want to try it out (https://www.lucidchart.com/).


Network Diagram from Lucidchart Tool

Geographical Data Center Requirements
Choosing a data center is a process that includes many requirements, but the location of the data center is often one that has the biggest impact. Following are some of the most important requirements for determining the geographic location of a data center:
- How close is the data center to the organization’s business offices and customers?
- What is the local climate?
- How geographically stable is the location?
- Is the location prone to extreme weather?
- How stable is the political climate for the area where the data center is located?
- What local IT talent is available?
- How stable is the infrastructure in the area, particularly power and network access?

Quiz:
1. What is the process of migrating from a standby system back to the standard active system called? A.Failback B.Site restore C.Reverse failover D. Return
2. To have a valid redundant resource, you need to replicate the active resource often enough to support your _____. A.RPO B.RTO C.SLA
3. Which standby site is typically the least costly overall? A.Hot B.Cold C.Warm
4. In disaster recovery a _____ is the step-by-step tasks that need to be performed to conduct the disaster recovery process. A.Playbook B.DR kit C.Site recovery plan D. Restore plan

Answers:
1. Failback
2. RPO
3. Cold
4. Playbook